← Search

Shuzhi Yu

2 accepted papers

2023

SemARFlow: Injecting Semantics into Unsupervised Optical Flow Estimation for Autonomous Driving

ICCV 2023poster

Unsupervised optical flow estimation is especially hard near occlusions and motion boundaries and in low-texture regions. We show that additional information such as semantics and domain knowledge can help better constrain this problem. We introduce SemARFlow, an unsupervised optical flow network de…

Cited by 8PDFcodeScholar
2022

Optical Flow Training under Limited Label Budget via Active Learning

ECCV 2022poster

"Supervised training of optical flow predictors generally yields better accuracy than unsupervised training. However, the improved performance comes at an often high annotation cost. Semi-supervised training trades off accuracy against annotation cost. We use a simple yet effective semi-supervised t…